Baixo ganho ponderal como preditor da retinopatia da prematuridade

• Ano de defesa: 2009
• Autor(a) principal: Fortes Filho, João Borges [UNIFESP]
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• dARK ID: ark:/48912/001300001wzmw
• Idioma: por
• Instituição de defesa: Universidade Federal de São Paulo (UNIFESP)
• Palavras-chave em Português: Retinopatia da prematuridade; Fatores de risco; Prevalência; Ganho ponderal
• Link de acesso: http://repositorio.unifesp.br/handle/11600/9137

Purposes: To analyze the low weight gain from birth to the sixth week of life as an independent risk factor for development of retinopathy of prematurity (ROP) as well as to evaluate a possible inverse relationship of weight gain with the stage of ROP. Methods: An institutional cohort, observational, and prospective study comparing incidence of ROP and weight gain after preterm birth. All infants with birth weight 1,500 grams and gestational age 32 weeks at birth between October 2002 and December 2006 that survived from the sixth to the 42th week of postmenstrual age were included. Principal clinical outcome were development of ROP in any stage and development of severe ROP in need of treatment. The main variable was the proportion of weight gain to the birth weight measured at the sixth week of life. To determine if the weight gain was influenced by the birth weight, patients were divided in two groups: Group 1 with birth weight 1,000 grams and Group 2 with birth weight >1,000 grams. Prevalence of ROP was determined in both groups. Chi-Square and Student - t test were used to compare patients with and without ROP. Logistic regressions were performed to determine if the weight gain was related to the development of ROP independent of other variables. Relative risk for ROP was calculated with 95% confidence interval. The accuracy of weight gain for the development of ROP in any stage or severe ROP were evaluated by receiver operating characteristic (ROC) curves with respective sensitivity and specificity cut offs and both positive and negative predictive values. Results: Of 317 studied preterms, 98 (30.9%) developed ROP in any stage. The mean weight gain among patients without ROP was 678.8 grams (DP 258.6) and among patients that developed the disease was 462.8 grams (DP 209.4), (P<0,001). The mean proportion of WG has reduced significantly in children that did not developed ROP to children that developed ROP stage 1 and 2 (P<0.001). It was not possible to determine this reduction in cases of ROP stage 3 or more. Adjusted logistic regression in Group 1 indicated 1.055 OR for ROP (CI95%: 1.028-1.083; P<0.001) and Group 2, 1.031 (CI95%; 1.008-1.054; P=0.007). The area under the ROC curve was 0.67% (CI95%: 0.598-0.729; P<0.001). The cutoff of 51,2% of the WG has shown 61.2% of sensitivity (CI95%: 51.3-70.5%), 64.4% of specificity (IC95%: 57.9-70.5%), 43.5% of positive predictive value for any stage of ROP (IC95%: 35.4-51.8%) and 78.8% of negative predictive value (IC95%: 72.3-84.3%). For severe ROP, the area under the curve was 0.63% (IC95%: 0.495-0.761; P=0.037), the cutoff has shown 62.5% of sensitivity (IC95%: 42.2-80.0%), 58.0% of specificity (IC95%: 52.3-63.6%), 10.8% of positive predictive value (IC95%: 6.5- 16.9%), and 95.0% negative predictive value (IC95%: 91.0-97.5%). Conclusions: Low weight gain measured by the 6th week of life was an important and independent risk factor and was capable to predict the development of ROP in any stage and severe ROP. It was not possible to show an inverse relationship between the weight gain and more severe stages of ROP due to a low number of these cases.